PharmaFormer predicts clinical drug responses through transfer learning guided by patient derived organoid

Abstract A major challenge in effective cancer treatment is the variability of drug responses among patients. Patient-derived organoids greatly preserve the genetic and histological characteristics even the drug sensitivities of primary tumor tissues, therefore provide a compelling approach to predi...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuru Zhou, Quanhui Dai, Yanming Xu, Shuang Wu, Minzhang Cheng, Bing Zhao
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:npj Precision Oncology
Online Access:https://doi.org/10.1038/s41698-025-01082-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849387817630695424
author Yuru Zhou
Quanhui Dai
Yanming Xu
Shuang Wu
Minzhang Cheng
Bing Zhao
author_facet Yuru Zhou
Quanhui Dai
Yanming Xu
Shuang Wu
Minzhang Cheng
Bing Zhao
author_sort Yuru Zhou
collection DOAJ
description Abstract A major challenge in effective cancer treatment is the variability of drug responses among patients. Patient-derived organoids greatly preserve the genetic and histological characteristics even the drug sensitivities of primary tumor tissues, therefore provide a compelling approach to predict clinical outcome. However, the individual organoid culture and following drug response test are time and cost-consuming, which hinders the potential clinical application. Here, we developed PharmaFormer, a clinical drug response prediction model based on custom Transformer architecture and transfer learning. PharmaFormer was initially pre-trained with the abundant gene expression and drug sensitivity data of 2D cell lines, and was then finalized through a model further fine-tuned with the limited organoid pharmacogenomic data accumulated at the present stage. Our results demonstrate that PharmaFormer, integrating both pan-cancer cell lines and organoids of a specific type of tumor, provides a dramatically improved accurate prediction of clinical drug response. This study highlights that advanced AI models combined with biomimetic organoid models will accelerate precision medicine and future drug development.
format Article
id doaj-art-012c99fac7f44b0a9045cf4ac1b835df
institution Kabale University
issn 2397-768X
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series npj Precision Oncology
spelling doaj-art-012c99fac7f44b0a9045cf4ac1b835df2025-08-20T03:42:29ZengNature Portfolionpj Precision Oncology2397-768X2025-08-019111110.1038/s41698-025-01082-6PharmaFormer predicts clinical drug responses through transfer learning guided by patient derived organoidYuru Zhou0Quanhui Dai1Yanming Xu2Shuang Wu3Minzhang Cheng4Bing Zhao5School of Basic Medical Sciences, Institute of Biomedical Innovation, The First Affiliated Hospital, Jiangxi Medical College, Nanchang UniversityZ Lab, bioGenous BIOTECHInstitute of Organoid Technology, Kunming Medical UniversityZ Lab, bioGenous BIOTECHSchool of Basic Medical Sciences, Institute of Biomedical Innovation, The First Affiliated Hospital, Jiangxi Medical College, Nanchang UniversitySchool of Basic Medical Sciences, Institute of Biomedical Innovation, The First Affiliated Hospital, Jiangxi Medical College, Nanchang UniversityAbstract A major challenge in effective cancer treatment is the variability of drug responses among patients. Patient-derived organoids greatly preserve the genetic and histological characteristics even the drug sensitivities of primary tumor tissues, therefore provide a compelling approach to predict clinical outcome. However, the individual organoid culture and following drug response test are time and cost-consuming, which hinders the potential clinical application. Here, we developed PharmaFormer, a clinical drug response prediction model based on custom Transformer architecture and transfer learning. PharmaFormer was initially pre-trained with the abundant gene expression and drug sensitivity data of 2D cell lines, and was then finalized through a model further fine-tuned with the limited organoid pharmacogenomic data accumulated at the present stage. Our results demonstrate that PharmaFormer, integrating both pan-cancer cell lines and organoids of a specific type of tumor, provides a dramatically improved accurate prediction of clinical drug response. This study highlights that advanced AI models combined with biomimetic organoid models will accelerate precision medicine and future drug development.https://doi.org/10.1038/s41698-025-01082-6
spellingShingle Yuru Zhou
Quanhui Dai
Yanming Xu
Shuang Wu
Minzhang Cheng
Bing Zhao
PharmaFormer predicts clinical drug responses through transfer learning guided by patient derived organoid
npj Precision Oncology
title PharmaFormer predicts clinical drug responses through transfer learning guided by patient derived organoid
title_full PharmaFormer predicts clinical drug responses through transfer learning guided by patient derived organoid
title_fullStr PharmaFormer predicts clinical drug responses through transfer learning guided by patient derived organoid
title_full_unstemmed PharmaFormer predicts clinical drug responses through transfer learning guided by patient derived organoid
title_short PharmaFormer predicts clinical drug responses through transfer learning guided by patient derived organoid
title_sort pharmaformer predicts clinical drug responses through transfer learning guided by patient derived organoid
url https://doi.org/10.1038/s41698-025-01082-6
work_keys_str_mv AT yuruzhou pharmaformerpredictsclinicaldrugresponsesthroughtransferlearningguidedbypatientderivedorganoid
AT quanhuidai pharmaformerpredictsclinicaldrugresponsesthroughtransferlearningguidedbypatientderivedorganoid
AT yanmingxu pharmaformerpredictsclinicaldrugresponsesthroughtransferlearningguidedbypatientderivedorganoid
AT shuangwu pharmaformerpredictsclinicaldrugresponsesthroughtransferlearningguidedbypatientderivedorganoid
AT minzhangcheng pharmaformerpredictsclinicaldrugresponsesthroughtransferlearningguidedbypatientderivedorganoid
AT bingzhao pharmaformerpredictsclinicaldrugresponsesthroughtransferlearningguidedbypatientderivedorganoid